Airborne Hyperspectral Data Compression
ABSTRACT: Operational airborne hyperspectral imagery (HSI) offers numerous advantages over traditional single or multiband sensing modalities for DoD intelligence, surveillance, and reconnaissance tasks. The extremely high information content, which captures unique spectral signatures of targets and materials of interest, can lessen the reliance on expert data analysts and improve response time through automated processing. However, the enormous stream of data collected with continuously operating HSI sensors far exceeds the throughput of existing real-time data analysis and telemetry systems. This SBIR program addresses the critical HSI data-processing needs of airborne platforms by developing algorithms and software for efficient, high-fidelity compression and storage that can be implemented in real time or near-real time, thereby eliminating existing bottlenecks that limit the timely utilization of these valuable data. Our compression system features lossless algorithms for data archival along with an innovative lossy method capable of producing compression ratios of order 50-fold. In Phase II, the software products will be implemented and tested with onboard processing hardware, for use with ACES-HY and similar hyperspectral sensing platforms. Structured file formats compliant with the National Imagery Transmission Format Standard (NITFS) will be developed to serve as a standardized exchange format. BENEFIT: The product of the proposed SBIR effort, after Phase II, will be a data compression algorithm software suite for immediate use in operational DoD and commercial hyperspectral systems and software packages for data analysis. The software will enable efficient telemetry of the imagery to the ground, thereby overcoming existing bottlenecks that limit the timely utilization of these data and enabling efficient data archiving both on-board and on ground computing systems. Sensors include ACES-HY and the many other military, civilian and commercial hyperspectral sensors that are or will be transmitting data from space or air to ground over the next two decades. Potential applications are wide-ranging and include military operations, emergency and disaster management, environmental monitoring, medicine, and manufacturing process control.
Small Business Information at Submission:
Spectral Sciences, Inc.
4 Fourth Avenue Burlington, MA -
Number of Employees: